We compared the Swedish Coronary Angiography and Angioplasty Registry with the Swedish 'Hospital Discharge Register' to assess contrast media (CM)-induced renal failure. Hospitals used only one type CM. From 2000 to 2003, iodixanol (iso-osmolar) was used in 45 485 patients, ioxaglate (low osmolar) in 12 440 subjects. To include the earlier used CM iohexol (low osmolar), analysis extended back to 1990 (86 334 patients). Incidence of clinically significant renal failure was greatest for patients receiving the iso-osmolar CM iodixanol (1.7%). Ioxaglate-treated patients had a significantly lower renal failure incidence (0.8%, P<0.001). The odds ratio for iodixanol-treated patients was significantly higher than for ioxaglate (1 vs 0.48, P<0.001). In subsets of either diabetic patients or patients with previous renal failure, odds ratios for renal failure remained greater in the iodixanol groups (P<0.01). Hospitals switching CM to iodixanol experienced a doubling in clinically significant renal failure after cardiac procedures. Dialysis was required in 0.2% of patients receiving iodixanol, which was significantly higher (P<0.01) than for ioxaglate-treated patients (0.1%). Iohexol-treated patients had a similar low risk for developing clinically significant renal failure (0.9%) as ioxaglate. In conclusion, risk of developing renal failure and required dialysis after coronary procedures is higher when patients received iodixanol than ioxaglate or iohexol.